文章基本信息

标题：A STUDY ON MINING FUZZY CLASSIFICATION RULES WITH EXCEPTIONS
本地全文：下载
作者：Somen Debnath ; Amarnath Pathak
期刊名称：Indian Journal of Computer Science and Engineering
印刷版ISSN：2231-3850
电子版ISSN：0976-5166
出版年度：2017
卷号：8
期号：2
页码：84-87
出版社：Engg Journals Publications
摘要：Now a days, searching of specific type of knowledge from the usual standards is very useful inseveral domains such as medical diagnosis, fraud detection , network traffic anomalies, economic analysisetc. Fuzzy association rules have been developed as a powerful tool for dealing with imprecision indatabases and offering a comprehensive representation of found knowledge. Adding fuzziness to normalclassification rules enable the rules to adapt to the real life decision making process. Besides, it also addsto the classification accuracy of obtained model and the rules look more accurate and reasonable. Furtherimprovement in classification accuracy can be achieved by discovering exceptions corresponding to thesefuzzy rules. Fuzzy rules augmented with exceptions (censors) are termed as Fuzzy CensoredClassification Rules (FCCRs) and such kinds of rules are best at handling uncertainties like vaguenessand ambiguity. These rules, being very efficient, have been widely used under exceptional circumstances.In this paper we have investigated all the algorithms used in past for discovering FCCRs. Based on reviewof literature, we have find out drawbacks and future direction with various issues.
关键词：CPR; Fuzzy Rules; FCCRs; Nature Inspired Algorithms.